Skip to main content
PMC home page
Drug and Alcohol Dependence Reports logoLink to Drug and Alcohol Dependence Reports
. 2024 Nov 15;13:100299. doi: 10.1016/j.dadr.2024.100299

The power of lived experience in optimizing US policymakers’ engagement with substance use research: A series of rapid-cycle randomized controlled trials

Elizabeth C Long a,⁎, Riley Loria b, Jessica Pugel a, Patrick O’Neill a, Camille C Cioffi c, Charleen Hsuan d, Glenn Sterner e, D Max Crowley a, J Taylor Scott a
PMCID: PMC11621502  PMID: 39650743

Abstract

Background

Research can inform policies on substance use/substance use disorders (SU/SUDs), yet there is limited experimental investigation into strategies for optimizing policymakers’ engagement with SU/SUD research. This study tested the use of narratives to boost policymakers’ research engagement.

Methods

In five rapid-cycle randomized controlled trials, SU/SUD research fact sheets were emailed to US legislative policymakers. We tested the use of narratives on the number of email opens, fact sheet clicks, and replies, relative to control emails without narratives. Narratives described lived experience with SU/SUD or motivations to study SU/SUD. The sender was a person with lived experience who authored the narrative or an author of the fact sheet.

Results

When the narrative was about the sender’s own lived experience (Trial 1), or when the narrative was about the sender's motivations to study SU/SUDs (Trial 2), the fact sheet was clicked more than the control (p=.049; p=.012; respectively). When the narrative was about someone else’s experience (Trials 3 and 4), the email was opened (p’s<.001) and replied to (p’s<.001) less, and the fact sheet was clicked (p’s<.001) less. Lastly, emails with lived experience narratives were replied to more than the control, regardless of sender (fact sheet author: p=.028; narrative author: p=.002; Trial 5), but were opened more if the sender authored the narrative (p<.001).

Conclusions

Policymakers’ engagement with SU/SUD research generally increased when the sender was telling their own story. This work highlights the power of people with lived experience and informs strategies for optimizing policymakers’ engagement with SU/SUD research.

Keywords: Drug policy, Substance use disorders, Science-policy communication, Research engagement

Highlights

  • •

    Increasing policymakers’ engagement with substance use research is challenging.

  • •

    Narratives as a best practice for science communication are rarely evaluated.

  • •

    Five trials show the benefit of narratives describing lived experience is nuanced.

  • •

    Engagement generally increased when the sender was the narrative author.

  • •

    This highlights the power of people with lived experience sharing their stories.

1. Introduction

Alcohol and drug misuse have contributed to a decline in life expectancy (Case and Deaton, 2015; Woolf et al., 2018), a reversal not seen in any other wealthy country, even prior to the COVID-19 pandemic. In 2021, more than 100,000 Americans died from drug overdoses (Centers for Disease Control and Prevention, 2021), and approximately 125,200 die from alcohol-attributable causes per year (Esser et al., 2022). These record-breaking rates, in addition to the high prevalence of substance use disorders (SUDs) (Mojtabai, 2022), come despite hundreds of research interventions designed to prevent, treat, or reduce the harms associated with substance use (SU) and SUDs. This paradox may stem from known barriers SU/SUD researchers face in getting rigorous and unbiased research into the hands of policymakers (Bogenschneider, 2020, Bogenschneider and Corbett, 2011, Long et al., 2021, Oliver et al., 2014, Scott et al., 2023a, Oliver et al., 2022), such as time constraints, university reward structures, lack of training opportunities for researchers to gain knowledge of the process of sharing research with policymakers, and differing norms and objectives between researchers and policymakers.

In the absence of rigorous and unbiased research effectively reaching policymakers, industry research efforts and lobbyists fulfill the demand for policy-relevant research evidence (McCambridge et al., 2018, Rossow and McCambridge, 2019). Such research may be biased since it is meant to skew policymakers towards a more favorable view of their cause (McGarity and Wagner, 2010). Though there are many instances of corporations performing high quality and unbiased research, there also is greater motivation in for-profit organizations to manipulate study designs in their industry’s favor (Lexchin et al., 2003), or simply to under report findings that might negatively impact their interests (Babor and Robaina, 2013). If biased research like this is most accessible to policymakers due to corporate financial backing, evidence-based policy changes that lead to public health benefits may be fruitless.

Alternatively, theory suggests that if unbiased, rigorous, and relevant research reaches the right policymakers at the right time in the right way, they may be more likely to use high quality evidence to inform policy decisions that can achieve population-level impacts (Bogenschneider and Corbett, 2021, Kingdon, 2011, Scott et al., 2014). For instance, the Icelandic Model of Adolescent Substance Use Prevention substantially reduced adolescent SU by increasing policymakers’ engagement with research on risk and protective factors through researcher-policymaker collaborations (Kristjansson et al., 2010, Sigfusdottir et al., 2010, Sigfúsdóttir et al., 2009). In the US, the Mainstreaming Addiction Treatment (MAT) Act of 2021 (Mainstreaming Addiction Treatment Act of 2021, n.d.) eliminated the waiver that was required through the Drug Enforcement Administration to prescribe buprenorphine, which was a restriction that was not supported by scientific evidence (National Academies of Sciences, 2019). These evidence-based provisions can increase access to treatment for opioid use disorder and save lives from overdoses.

Researchers and science communicators can increase the successful use of quality scientific evidence in policymaking by improving policymakers’ access to research evidence on SU/SUD through relationships with them. The SciComm Optimizer for Policy Engagement (SCOPE) model aims to facilitate engagement between researchers and policymakers via research disseminations by using continuous quality improvement to experimentally test different message frames (see Fig. 1). The SCOPE model leverages theory pointing to the need for research evidence to be relevant to current policy priorities and available in real time so that it can be used within discrete policy windows (Mackie et al., 2015). In fact, one of the most frequently reported facilitators of policymakers’ use of research evidence involves access to relevant and timely research (Oliver et al., 2014). This contrasts with one-way dissemination efforts that “push” research information without considering policymakers’ current policy priorities (Tseng, 2012).

Fig. 1.

Fig. 1

Open in a new tab

The SciComm Optimizer for Policy Engagement (SCOPE) Model. Note. CQI = continuous quality improvement.

Policymakers’ current policy goals can be learned through interactions occurring as part of relational research-policy bridging approaches, such as the Research-to-Policy Collaboration (RPC) model, or through scanning their public discourse (Crowley et al., 2018, Scott et al., 2019, Crowley et al., 2021b, Crowley et al., 2021a). The RPC model aims to build relationships between researchers and policymakers by assessing their needs for research evidence and responding accordingly with nonpartisan research information. If several different legislative staff have needs and questions about the same topic or reflect similar priorities, it would be considered timely and relevant enough to respond with broad dissemination of corresponding research syntheses, typically in the form of one- to two-page fact sheets. This dissemination of timely and relevant research evidence can in turn facilitate additional interactions. Accordingly, the SCOPE model supplements this interactive bridging approach by forming feedback loops and expanding the number of policymakers who receive the research syntheses resulting from the interactions. More specifically, the steps of model include: assessing policymakers’ current policy priorities, developing corresponding content in the form of concise fact sheets, identifying officials and/or staff who are likely to find the content useful, creating different messaging frames to test, disseminating the fact sheets, and analyzing the results so they may inform hypotheses for future testing.

Previous work using the SCOPE model has shown that evoking emotional responses (Long et al., 2021) and cueing a topic’s relevance (Scott et al., 2023a)can increase policymakers’ electronic engagement with research evidence. Further, a recent experimental evaluation of the SCOPE model’s impact demonstrated its ability to increase state legislators’ use of research evidence in social media posts (Scott et al., 2023b). The effectiveness of these strategies for improving engagement and use of research demonstrates the utility of tailoring communication techniques to increase the psychological closeness and personal relevance of an issue. Psychological distance in this context represents how close the reader feels to the subject of a message (Liberman and Trope, 2014). These SCOPE studies, however, were focused on research topics related to racial justice and the COVID-19 pandemic, leaving strategies for optimizing engagement with SU/SUD research unclear.

Related work has shown that framing effects can influence behaviors and beliefs. For example, cross-sectional research finds that state legislators’ support for opioid use disorder and behavioral health parity laws were more strongly associated with their personal beliefs about treatment and fiscal impacts rather than their political party, ideology, or individual demographics (Nelson and Purtle, 2020, Purtle et al., 2019). Additionally, varying narratives about particular topics can influence public support for mental health and SUD policy issues (Kennedy-Hendricks et al., 2016, McGinty et al., 2015, McGinty et al., 2018).

This suggests that framing strategies that increase personal relevance and decrease psychological distance from research messages could be effective in shifting lawmakers’ beliefs and later actions. Topics that feel psychologically closer feel more personally relevant. While many factors can influence psychological closeness, we focus on the use of narratives in the current study for several reasons. Narratives are especially useful in engaging communication, with past work showing that narratives, especially first-person narratives, are effective in reducing psychological distance (Cao and Decker, 2015). In communications research, transportation theory describes how narratives can reduce the psychological distance of a topic. Being engrossed and invested in a story allows readers to internalize a message with which they might not normally engage deeply (Gebbers et al., 2017, Green and Brock, 2000). First-person narratives are particularly good at reducing distance as they naturally evoke perspective taking (Salem et al., 2017, Van Lissa et al., 2016). Accordingly, using narratives may be an effective strategy for improving policymakers’ engagement with SU/SUD research.

However, many persuasion/advocacy techniques, including narratives, have yet to be tested in the unique context of messages to policymakers. Additionally, there is a great deal of competition for policymakers’ attention. Since lawmakers have a particularly powerful and direct influence on SU/SUD policies, it is vital to investigate framing strategies to optimize research engagement efforts with this population. The aim of the present study is to experimentally test different narrative frames as tools for optimizing policymakers’ engagement with SU/SUD research. We hypothesized that using narratives in the email bodies would result in more email opens, clicks on the fact sheet links, and email replies than not using narratives. We generated more specific hypotheses between each trial based on the prior trial’s results, predicting that lived experience with SUD, first-person narratives, shorter narratives, and the sender being the narrative author would generate greater engagement compared to the control. These predictions are all driven by an overarching theory that increasing psychological closeness should increase engagement.

2. Methods

2.1. Study design and participants

Following the steps outlined in the SCOPE model, five rapid-cycle randomized controlled trials were conducted between May 2020 and August 2022. In each trial, a fact sheet that contained timely research on SU/SUD (as identified through RPC meetings or public discourse scans) was disseminated via email to US state and federal policymakers who work on committees related to substance use and their staff (Ns = 5964 – 11,212). Participants were identified via an online Client Relationship Management (CRM) software, Quorum, which was developed specifically for legislative outreach. They were then randomized into different messaging conditions testing the use of narratives. In accordance with the nature of rapid-cycle experimentation, results of trials informed the hypotheses for subsequent trials. This study was deemed to be exempt by the Institutional Review Board at the author’s university because it is a low-risk study involving observation of natural, routine behavior that elected public officials engage in daily (i.e., engaging with an email). However, we added a disclosure statement to the email, “Data on email open and click rates may be used for research and quality improvement.” Contact information was also provided so the email recipient could contact study personnel if desired.

Due to the observational nature of this project, we do not have access to demographic information. However, as a whole, state legislators are predominantly White (82 %), male (76 %), and on average 56 years old (Kurtz, 2015). Similarly, Congressional legislators are primarily White (80.3 %) and male (75.8 %; Quorum, 2018).

2.2. Procedures

Quorum was used to select legislators and staff, send the messages, and monitor email opens and fact sheet clicks. Emails were created to appear typical for an informational email, written in default sized black text, using no special formatting or graphics. Additionally, the sender name appeared as their first name and last name only (i.e., degrees were not mentioned), and the email was sent from a domain independent of the author’s affiliation. Recipients could determine affiliation and credentials from the sender’s signature block in the email. Recipients were invited to reply if they had questions or wanted to discuss further.

2.2.1. Trial specifics

Five messaging trials were conducted by comparing various forms of narratives to a control message with no narrative that only described the topic of the fact sheet. All emails within a trial included a link to the same current and relevant fact sheet about SU/SUD. Each email in a given trial used the same subject line and had similar lengths (except Trial 4), while subject lines between trials differed to reflect the topic of the fact sheet that was being disseminated. Only the email body and sender were manipulated. See the Supplement for the specific test, subject line, sender, and full email body content for each trial. In each trial, message recipients were randomly assigned to message conditions.

In Trials 1 and 2, the narratives were written by a person with lived experience of addiction and recovery at the time of disseminating the research, not by the research team. We aimed to empower the narrative authors to authentically tell their stories how they wished, include aspects of their identity that they deemed to be important, and to share as much or as little of their story as they felt comfortable doing so. We did not feel that it would be appropriate to edit these individuals’ own personal narratives. In Trials 3, 4, and 5, the narratives were pulled from the Story Powered Initiatives website (https://storypowered.org), which is an initiative that aims to reduce stigma by encouraging people to share their lived experience with SUD. All people who submit their stories to be shared on this website agree to have them shared publicly. These narratives were only adapted for the testing condition (i.e., the person it was told in and length). In Trials where the sender was the narrative author, we worked with the author for them to approve the messages before we sent them and got their approval to make the emails appear to come from them via Quorum, which allowed us to track the email metrics. Because the narratives were written by the people with lived experience themselves, some of the language in these narratives may include what the SU/SUD field classifies as stigmatizing, or they may include identity attributes that may not appear to be relevant to the testing (e.g., race). However, we want to emphasize that they were written how the person with lived experience wanted them to be and not necessarily how academic researchers may think it should be.

Trial 1 tested a control condition email against the use of a narrative in the email body that described the sender’s personal lived experience with SUD. The sender for both message conditions was the author of the narrative, who also contributed content to the fact sheet.

Trial 2 tested the control condition against two types of narratives, both from a SU/SUD researcher with lived experience who contributed to the fact sheet. The first narrative condition described the author’s research narrative as inspired by her lived experience with SUD. The second narrative condition described the author’s research narrative without reference to their lived experience of SUD.

Trial 3 tested the control condition against the narrative written in first person and the narrative written in third person. The sender was an editor of the fact sheet, and, notably, was not the narrative author.

Trial 4 tested the control condition against a long version of the narrative condition and an abbreviated version of the narrative condition. Both were written in third person. The sender was an author of the fact sheet, and again was not the narrative author.

Trial 5 tested the control condition against the narrative being told in the 3rd person, sent by one of the fact sheet authors, and the narrative being told in the 1st person, sent by the narrative author, with the goal of further examining the role of the sender’s identity in relationship to the narrative. In other words, we were interested in testing the effects of the sender telling someone else’s story and the sender telling their own story.

2.3. Outcomes

Email open rates, fact sheet clicks, and replies for all participants were tracked for 14 days after the emails were sent. Opens were tracked using an image download tracker, which is standard practice in the industry. The tracker involves an invisible, one-pixel image in the email body that is downloaded when the email is opened. While the subject lines were not experimentally tested, we expect that the beginning of the email body or preview and the sender may have an effect on email open rates, given that these characteristics will be visible to the recipient prior to opening. Although the amount of text that is visible in the preview differs by the recipient’s email provider, we expect that all recipients were able to view at least the greeting. However, we do acknowledge that in Trials 4 and 5, the beginning of the email body is identical across conditions. Industry-standard magic links to the fact sheets registered how many times they are clicked.

2.4. Statistical analysis

For each trial, the control condition was the reference to which the experimental conditions were compared. Negative binomial regressions were conducted to test whether the experimental manipulations resulted in higher rates of email opens and clicks on the fact sheet compared to a control email, given the overdispersion of the distribution. The probability of a reply to the email was analyzed using logistic regression, since the reply variable was a dichotomous variable (reply vs. no reply).

3. Results

Results on email opens and fact sheet clicks are presented in Table 1, Table 2, respectively. Results on email replies are presented in Table 3.

Table 1.

Results of negative binomial models testing the effects of email body on number of email opens.

N Opens (ŷ) % Opens IRR
(95 % CI) 		SE z p
Trial 1: Challenges faced by parents with addiction 5964 2675 (0.45) 23.5
 Control (ref); Narrative & fact sheet author sender 2974 1280 (0.43) 23.4 - - - -
 Personal narrative with lived experience; Narrative & fact sheet author sender 2991 1395 (0.47) 23.6 1.08
(0.96, 1.23)		 .07 1.26 .206
Trial 2: Substance use research for [legislator name] 9496 5726 (0.60) 26.0
 Control (ref); Narrative & fact sheet author sender 3190 1907 (0.60) 26.6 - - - -
 Research narrative with lived experience; Narrative & fact sheet author sender 3134 1819 (0.58) 26.1 0.97
(0.86, 1.10)		 0.06 −0.47 .636
 Research narrative with no lived experience; Narrative & fact sheet author sender 3172 2000 (0.63) 25.3 1.06
(0.93, 1.19)		 0.07 0.86 .388
Trial 3: Fach sheet on interventions for substance use disorders 11,099 5960 (0.54) 24.2
 Control (ref); Fact sheet editor sender 3703 3142 (0.85) 38.0 - - - -
 Third-person narrative; Fact sheet editor sender 3699 1456 (0.39) 17.7 0.46
(0.41, 0.52) 		0.03 −13.60 <.001
  First-person narrative; Fact sheet editor sender 3697 1362 (0.37) 16.7 0.43
(0.39, 0.49) 		0.02 −14.67 <.001
Trial 4: Resource on substance use helplines and peer support 11,212 8012 (0.71) 36.0
 Control (ref); Fact sheet author sender 3729 3093 (0.83) 36.2 - - - -
 Long third-person; Fact sheet author sender 3747 2404 (0.64) 35.7 0.77
(0.71, 0.84) 		0.03 −5.74 <001
 Short third-person; Fact sheet author sender 3736 2515 (0.67) 36.1 0.81
(0.74, 0.89) 		0.03 −4.68 <.001
Trial 5: Research on prescription opioid labels 10,604 6787 (0.64) 36.3
 Control (ref); Fact sheet author sender 3533 2128 (0.60) 35.9 - - - -
 Third-person; Fact sheet author sender 3534 2149 (0.61) 35.9 1.00
(0.93, 1.10)		 0.04 0.21 .830
 First-person; Narrative author sender 3537 2510 (0.71) 37.2 1.18
(1.08, 1.28) 		0.05 3.75 <.001
Open in a new tab

Note. ŷ = predicted value of opens (the mean); IRR = incident ratio interval; CI = confidence interval; SE = standard error. % opens represents the percentage of recipients who opened the email, while the raw number of opens is the total number of opens across all recipients.

Table 2.

Results of negative binomial models testing the effects of email body on number of clicks on the fact sheet.

N Fact Sheet Clicks (ŷ) % Clicks IRR
(95 % CI) 		SE z p
Trial 1: Challenges faced by parents with addiction 5964 454 (0.08) 4.2
 Control (ref); Narrative & fact sheet author sender 2974 192 (0.06) 3.4 - - - -
 Personal narrative with lived experience; Narrative & fact sheet author sender 2991 262 (0.09) 5.0 1.37
(1.00, 1.87) 		0.22 1.97 .049
Trial 2: Substance use research for [legislator name] 9496 1298 (0.14) 7.8
 Control (ref); Narrative & fact sheet author sender 3190 384 (0.12) 7.3 - - - -
 Research narrative with lived experience; Narrative & fact sheet author sender 3134 414 (0.13) 7.5 1.07
(0.86, 1.34)		 0.12 0.64 .523
 Research narrative with no lived experience; Narrative & fact sheet author sender 3172 500 (0.16) 8.4 1.32
(1.06, 1.64) 		0.15 2.51 .012
Trial 3: Fach sheet on interventions for substance use disorders 11,099 857 (.08)
 Control (ref); Fact sheet editor sender 3703 601 (.16) 16.2 - - - -
 Third-person narrative; Fact sheet editor sender 3699 140 (.04) 3.8 0.23
(0.19, 0.29) 		0.03 −12.83 <.001
 First-person narrative; Fact sheet editor sender 3697 116 (.03) 3.1 0.19
(0.15, 0.24) 		0.02 −13.71 <.001
Trial 4: Resource on substance use helplines and peer support 11,212 821 (.07)
 Control (ref); Fact sheet author sender 3729 479 (.13) 12.8 - - - -
 Long third-person; Fact sheet author sender 3747 168 (.04) 4.5 0.35
(0.27, 0.45) 		0.05 −8.00 <.001
 Short third-person; Fact sheet author sender 3736 174 (.05) 4.7 0.36
(0.28, 0.47) 		0.05 −7.76 <.001
Trial 5: Research on prescription opioid labels 10,604 502 (.05)
 Control (ref); Fact sheet author sender 3533 164 (.05) 4.6 - - - -
 Third-person; Fact sheet author sender 3534 143 (.04) 4.0 0.87
(0.64, 1.19)		 0.14 −0.86 .388
 First-person; Narrative author sender 3537 195 (.06) 5.5 1.19
(0.88, 1.60)		 0.18 1.12 .261
Open in a new tab

Note. ŷ = predicted value of clicks (the mean); IRR = incident ratio interval; CI = confidence interval; SE = standard error. % clicks represents the percentage of recipients who clicked the link, while the raw number of clicks is the total number of clicks across all recipients.

Table 3.

Results of logistic models testing the effects of email body on whether the recipient replied.

Replies (ŷ) OR
(95 % CI) 		SE z p
Trial 1: Challenges faced by parents with addiction 196 (0.03)
 Control (ref); Narrative & fact sheet author sender 96 (0.03) - - - -
 Personal narrative with lived experience; Narrative & fact sheet author sender 100 (0.03) 1.05
(0.79, 1.39)		 0.15 0.32 .746
Trial 2: Substance use research for [legislator name] 171 (0.02)
 Control (ref); Narrative & fact sheet author sender 65 (0.02) - - - -
 Research narrative with lived experience; Narrative & fact sheet author sender 44 (0.01) 0.67
(0.46, 0.99) 		0.13 −2.01 .044
 Research narrative with no lived experience; Narrative & fact sheet author sender 62 (0.02) 0.97
(0.69, 1.38)		 0.17 −0.14 .885
Trial 3 - N/A - - - - -
Trial 4: Resource on substance use helplines and peer support
 Control (ref); Fact sheet author sender 133 (.04) - - - -
 Long third-person; Fact sheet author sender 61 (.02) 0.45
(0.33, 0.61) 		0.07 −5.14 <.001
 Short third-person; Fact sheet author sender 74 (.02) 0.55
(0.41, 0.73) 		0.08 −4.11 <.001
Trial 5: Research on prescription opioid labels
 Control (ref); Fact sheet author sender 35 (.01) - - - -
 Third-person; Fact sheet author sender 56 (.02) 1.61
(1.05, 2.46) 		0.35 2.19 .028
 First-person; Narrative author sender 67 (.02) 1.93
(1.28, 2.91) 		0.41 3.13 .002
Open in a new tab

Note. ŷ = predicted value of replies (the mean); OR = odds ratio; CI = confidence interval; SE = standard error. Response data not available for Trial 3.

Trial 1 compared a personal narrative with lived experience to a control message with no lived experience, with the sender being the narrative author and one of the fact sheet authors. Recipients of the email body that described a personal narrative with lived SU/SUD experience clicked the fact sheet 1.37 times as frequently as the control that did not include a narrative (p=.049). There were, however, no significant differences in opens by email message condition and there were no differences in probability of a reply by condition.

Trial 2 compared the impact of a research narrative with and without lived experience to a control message, with the sender being the narrative author and one of the fact sheet authors. The recipients of the email body that described the research narrative with no lived experience clicked the fact sheet 1.32 times as frequently as the control (p=.012). The research narrative with lived experience was significantly less likely to be replied to than the control (OR=0.67, p=.044). There were no significant differences in opens for either narrative condition.

Trial 3 compared the impact of third-person and first-person narratives to a control message, with the sender being a fact sheet editor (not the narrative author). Unlike the prior trials, the control outperformed the narrative conditions in both clicks and opens. The third-person narrative resulted in about half (IRR=0.46) as many opens (p<.001) and roughly a fifth (IRR=0.23) as many clicks (p<.001) as the control. Similarly, the first-person narrative resulted in 0.43 times as many opens (p<.001) and only a fifth (OR=0.19) as many clicks as the control (p<.001). Due to a technical error, results are not available for replies in Trial 3.

Trial 4 compared short and long narratives, written in third person, to a control message, with the sender being a fact sheet author (not the narrative author). Those who received the long narrative opened the email 0.77 as many times (p<.001), clicked the fact sheet link 0.35 as many times (p<.001), and were about half (OR=0.45) as likely to have replied as recipients of the control message (p<.001). Those who received the short narrative opened the email 0.81 as many times (p<.001), clicked the fact sheet link 0.36 as many times (p<.001), and were about half (OR=0.55) as likely to have replied (p<.001) as recipients of the control message.

Trial 5 compared a first-person narrative, with the sender being the narrative author (i.e., the sender telling their own narrative) and a third-person narrative, with the sender being one of the fact sheet authors (i.e., the sender telling someone else’s narrative) to a control message without a narrative and sent from the fact sheet author. The third-person, fact sheet author email was 1.61 times more likely to have received a reply than the control (p=.028), but was not significantly different from the control in terms of email opens or fact sheet clicks. The first-person, narrative author email was opened 1.18 times as frequently (p<.001) and was nearly twice as likely to have received a response (OR=1.93, p=.002) than the control message.

4. Discussion

The present study aimed to test the use of narratives and sender effects (i.e., fact sheet author/editor vs. narrative author) on state and federal policymakers’ engagement with SU/SUD research. The results were nuanced and our hypothesis that using narratives would improve policymakers’ engagement with SU/SUD was only supported under certain conditions (see Table 4). At times, using narratives even backfired. First, whether the use of narratives in the email body is an effective strategy for optimizing engagement may depend on the sender. Specifically, narratives seem to be particularly effective when they are sent from the same person telling the story, as opposed to someone else telling their story. This highlights the potential power of people with lived experience in telling their story.

Table 4.

Results summary across all trials.

Opens Clicks Replies
Trial 1: Challenges faced by parents with addiction
 Control (ref); Narrative & fact sheet author sender
 Personal narrative with lived experience; Narrative & fact sheet author sender - ↑ -
Trial 2: Substance use research for [legislator name]
 Control (ref); Narrative & fact sheet author sender
 Research narrative with lived experience; Narrative & fact sheet author sender - - ↓
 Research narrative with no lived experience; Narrative & fact sheet author sender - ↑ -
Trial 3: Fach sheet on interventions for substance use disorders
 Control (ref); Fact sheet editor sender
 Third-person narrative; Fact sheet editor sender ↓ ↓ n/a
 First-person narrative; Fact sheet editor sender ↓ ↓ n/a
Trial 4: Resource on substance use helplines and peer support
 Control (ref); Fact sheet author sender
 Long third-person; Fact sheet author sender ↓ ↓ ↓
 Short third-person; Fact sheet author sender ↓ ↓ ↓
Trial 5: Research on prescription opioid labels
 Control (ref); Fact sheet author sender
Third-person; Fact sheet author sender - - ↑
 First-person; Narrative author sender ↑ - ↑
Open in a new tab

Note. Arrows indicate statistically significant effects; - indicates non-significance.

Second, the findings from Trial 2 suggest that narratives were beneficial when the sender’s identity appears to be congruent with a person’s background, as opposed to when the story is from someone with multiple relevant identities that require the recipient to reconcile. For example, a researcher with lived experience of SUDs may be difficult to reconcile because of the stigmatization of people with SUD. These complex identities may make reducing psychological distance more challenging. Future work, however, is needed to further explore these differences in engagement.

When the sender shared someone else’s narrative, narratives did not improve engagement and sometimes hindered it, regardless of whether the narrative was written in first or third person or length of the narrative. Though we expected first-person narratives to be more effective, these results suggest that direct engagement with the actual narrative author may be necessary for narratives to increase psychological closeness. Adding the additional step of someone else’s narrative may make the narrative seem more contrived and make the recipient less inclined to engage. If the value of narratives as a messaging tool lies in its ability to connect the reader and its author, then it follows that an added layer of distance in the form of a third-party sender would dampen its effects. Collectively, it seems that being authentic and transparent is effective for reducing psychological distance and optimizing policymakers’ engagement with research. However, given that this work is in its infancy, further trials are needed to investigate alternative explanations.

Our findings are somewhat consistent with the broader literature showing that framing can influence behaviors and beliefs (Kennedy-Hendricks et al., 2016, McGinty et al., 2015, Nelson and Purtle, 2020, Purtle et al., 2019, McGinty et al., 2018). More specifically, McGinty and colleagues (McGinty et al., 2015) found that portraying successful treatment of mental disorders and SUDs improves public support and reduces stigma, which is consistent with our findings showing that the fact sheets in emails with narratives from individuals who have had success with treatment and are in active recovery were clicked on more than the control emails.

While the current study focused on testing narratives within the email body rather than subject lines, our prior work focused on testing whether different subject lines influenced email open rates. We have shown, for instance, that using words that evoke emotional responses in the subject lines of emails disseminating racial justice research and cueing relevance with personalized subject lines of emails disseminating research that was timely during the COVID-19 pandemic increased the likelihood of policymakers’ opening the email (Long et al., 2021, Scott et al., 2023a). Future work is needed, however, to investigate whether these findings generalize to emails disseminating SU research and whether certain subject lines increase the open rates of emails with narratives of lived experience more than others.

4.1. Limitations

Our results should be considered in the context of several limitations. First, it is unclear whether our results will replicate in other contexts, issue areas, senders, and political climates. Similarly, results may not generalize to policymakers who work on issues that are not related to SU or to policymakers outside of the US. Second, though emails within a trial were sent simultaneously, emails across trials were sent on different days and times, which may have affected the results due to the current political and environmental context (e.g., perceptions of science, prevalence of SUDs).

Third, because the personal narratives were not written by the research team, we had less control over other potential influences in the email bodies that may have affected policymakers’ engagement with the research. We were not able to hold certain variables constant that we would have been able to otherwise if we had written the narratives ourselves. While it may be interesting for researchers to manipulate narratives to identify what facets of a story are most impactful for policymakers, there is risk associated with this. We might unintentionally contribute to the societal discourse that certain people’s stories matter more than others, disempowering some voices. There are several resources available for researchers and organizations to better understand how to facilitate the sharing of lived experience in a way that is responsible and protects against unintended consequences. These include a foundational booklet from the Canadian HIV/AIDS Legal Network (2006), a set of guidelines from the National Council for Mental Wellbeing (Plante and Lovell, 2024), commentary from Cioffi et al. (2024) who expand upon these guidelines, and a number of recent academic manuscripts (Cioffi et al., 2023, Crabtree et al., 2016, Damon et al., 2017, Lipsett et al., 2023).

Fourth, while our engagement rates were overall low, they are consistent with previous outreach efforts with policymakers (~20 %-40 % open rates, ~3 %-8 % click rates; Long et al., 2021; Scott et al., 2023a, Scott et al., 2023b). Finally, the same legislative office could have been emailed for more than one of the trials, which could have influenced engagement with emails later in the trial. However, with significant staff turnover, it is entirely possible that the same staffer did not see all five emails. Further, the sender between trials (and within Trial 5) differed, the specific topic for each trial differed, and the emails were sent with months in between. Accordingly, we do not expect that recipients would have realized the emails were related.

4.2. Conclusions

This work helps to inform strategies for increasing policymakers’ electronic engagement with SU/SUD research by demonstrating that the benefit of using narratives for this purpose is nuanced. In general, policymakers were more likely to engage with research when the narrative was about the sender, and when the sender’s identity appeared to be congruent with the person’s background, but not when the sender was the fact sheet author or editor. This work highlights the power of people with lived experience sharing their stories in optimizing policymakers’ engagement with SU/SUD research, which is critical for improving the use of rigorous, unbiased scientific research in SU/SUD policies and improving public health outcomes.

Funding

Generous support for this work was provided by the William T. Grant Foundation (grant 200884), the National Institute on Drug Abuse (R01DA056627), the Eunice Kennedy Shriver National Institute of Child Health and Human Development (grant P50HD089922), the National Science Foundation (grant 2030660), and the following Institutes from the Pennsylvania State University: the Huck Institutes of the Life Sciences, the Social Science Research Institute, and the Consortium for Substance Use and Addiction.

CRediT authorship contribution statement

Jessica Pugel: Writing – review & editing, Investigation, Formal analysis, Data curation. Riley Loria: Writing – review & editing, Visualization, Formal analysis. Elizabeth Long: Writing – review & editing, Writing – original draft, Methodology, Funding acquisition, Conceptualization. J. Taylor Scott: Writing – review & editing, Supervision, Funding acquisition, Conceptualization. D. Max Crowley: Supervision, Funding acquisition. Glenn Sterner: Resources, Funding acquisition, Conceptualization. Charleen Hsuan: Writing – review & editing, Funding acquisition, Conceptualization. Camille C. Cioffi: Writing – review & editing, Resources. Patrick O’Neill: Writing – review & editing, Data curation.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

We thank Research-to-Policy Collaboration Rapid Response Network researchers and fact sheet authors who authorized us to distribute their work under their names and expertly navigated the responses from policymakers.

Footnotes

Appendix A

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.dadr.2024.100299.

Appendix A. Supplementary material

Supplementary material

mmc1.docx (33.1KB, docx)

References

  1. Babor T.F., Robaina K. Public health, academic medicine, and the alcohol industry’s corporate social responsibility activities. Am. J. Public Health. 2013;103:206–214. doi: 10.2105/AJPH.2012.300847. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bogenschneider K. Positioning universities as honest knowledge brokers: best practices for communicating research to policymakers. Fam. Relat. 2020;69:628–643. doi: 10.1111/fare.12339. [DOI] [Google Scholar]
  3. Bogenschneider K., Corbett T. Evidence-based Policymaking: Envisioning a New Era of Theory, Research, and Practice. 2nd ed. Taylor & Francis Group; New York, NY: 2021. (ed) [Google Scholar]
  4. Bogenschneider K., Corbett T.J. Routledge; 2011. Evidence-based policymaking: insights from policy-minded researchers and research-minded policymakers. [Google Scholar]
  5. Canadian HIV/AIDS Legal Network. “Nothing About Us Without Us”: Greater, Meaningful Involvement of People Who Use Illegal Drugs: A Public Health, Ethical, and Human Rights Imperative (Booklet), 2006.
  6. Cao X., Decker D. Psychological distancing: the effects of narrative perspectives and levels of access to a victim’s inner world on victim blame and helping intention. Int. J. Nonprofit Volunt. Sect. Mark. 2015;20:12–24. [Google Scholar]
  7. Case A., Deaton A. Rising morbidity and mortality in midlife among white non-Hispanic Americans in the 21st century. Proc. Natl. Acad. Sci. 2015;112:15078–15083. doi: 10.1073/pnas.1518393112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Centers for Disease Control and Prevention, 2021. Drug overdose deaths in the US top 100,000 annually [WWW Document]. URL 〈https://www.cdc.gov/nchs/pressroom/nchs_press_releases/2021/20211117.htm〉
  9. Cioffi C.C., Hibbard P.F., Hagaman A., Tillson M., Vest N. Perspectives of researchers with lived experience in implementation science research: opportunities to close the research-to-practice gap in substance use systems of care. Implement. Res. Pract. 2023;4 doi: 10.1177/26334895231180635. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Cioffi C.C., Flinn R.E., Pasman E., Gannon K., Gold D., McCabe S.E., Kepner W., Tillson M., Colditz J.B., Smith D.C., Bohler R.M., O’Donnell J.E., Hildebran C., Montgomery B.W., Clingan S., Lofaro R.J. Beyond the 5-year recovery mark: perspectives of researchers with lived and living experience on public engagement and discourse. Int. J. Drug Policy. 2024;133 doi: 10.1016/j.drugpo.2024.104599. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Crabtree A., Latham N., Bird L., Buxton J. Results of a participatory needs assessment demonstrate an opportunity to involve people who use alcohol in drug user activism and harm reduction. Harm. Reduct. J. 2016;13:37. doi: 10.1186/s12954-016-0126-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Crowley D., Scott T., Fishbein D. Translating prevention research for evidence-based policymaking: results from the research-to-policy collaboration pilot. Prev. Sci. 2018;19:260–270. doi: 10.1007/s11121-017-0833-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Crowley D.Max, Scott J.T., Long E.C., Green L., Israel A., Supplee L., Jordan E., Oliver K., Guillot-Wright S., Gay B., Storace R., Torres-Mackie N., Murphy Y., Donnay S., Reardanz J., Smith R., McGuire K., Baker E., Antonopoulos A., McCauley M., Giray C. Lawmakers’ use of scientific evidence can be improved. Proc. Natl. Acad. Sci. 2021;118 doi: 10.1073/pnas.2012955118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Crowley D.Max, Scott J.T., Long E.C., Green L., Giray C., Gay B., Israel A., Storace R., McCauley M., Donovan M. Cultivating researcher-policymaker partnerships: a randomized controlled trial of a model for training public psychologists. Am. Psychol. 2021;76:1307–1322. doi: 10.1037/amp0000880. 〈https://psycnet.apa.org/doi/10.1037/amp0000880〉 [DOI] [PubMed] [Google Scholar]
  15. Damon W., Callon C., Wiebe L., Small W., Kerr T., McNeil R. Community-based participatory research in a heavily researched inner city neighbourhood: perspectives of people who use drugs on their experiences as peer researchers. Soc. Sci. Med. 2017;176:85–92. doi: 10.1016/j.socscimed.2017.01.027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Esser M.B., Sherk A., Subbaraman M.S., Martinez P., Karriker-Jaffe K.J., Sacks J.J., Naimi T.S. Improving estimates of alcohol-attributable deaths in the United States: impact of adjusting for the underreporting of alcohol consumption. J. Stud. Alcohol Drugs. 2022;83:134–144. doi: 10.15288/jsad.2022.83.134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gebbers T., De Wit J.B., Appel M. Health communication| transportation into narrative worlds and the motivation to change health-related behavior. Int. J. Commun. 2017;11:21. [Google Scholar]
  18. Green M.C., Brock T.C. The role of transportation in the persuasiveness of public narratives. J. Pers. Soc. Psychol. 2000;79:701. doi: 10.1037//0022-3514.79.5.701. [DOI] [PubMed] [Google Scholar]
  19. Kennedy-Hendricks A., McGinty E.E., Barry C.L. Effects of competing narratives on public perceptions of opioid pain reliever addiction during pregnancy. J. Health Polit. Policy Law. 2016;41:873–916. doi: 10.1215/03616878-3632230. [DOI] [PubMed] [Google Scholar]
  20. Kingdon, J., 2011. Agendas, Alternatives, and Public Policies, Update Edition, with an Epilogue on Health Care, 2nd Edition [WWW Document]. URL 〈https://www.pearson.com/content/one-dot-com/one-dot-com/us/en/higher-education/program.html〉 (accessed 4.2.22).
  21. Kristjansson A.L., James J.E., Allegrante J.P., Sigfusdottir I.D., Helgason A.R. Adolescent substance use, parental monitoring, and leisure-time activities: 12-year outcomes of primary prevention in Iceland. Prev. Med. 2010;51:168–171. doi: 10.1016/j.ypmed.2010.05.001. [DOI] [PubMed] [Google Scholar]
  22. Kurtz, K., 2015. Who we elect: The demographics of state legislatures. National Conference of State Legislatures.
  23. Lexchin J., Bero L., Djulbegovic B., Clark O. Pharmaceutical industry sponsorship and research outcome and quality: systematic review. BMJ. 2003;326:1167–1170. doi: 10.1136/bmj.326.7400.1167. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Liberman N., Trope Y. Traversing psychological distance. Trends Cogn. Sci. 2014;18:364–369. doi: 10.1016/j.tics.2014.03.001. [DOI] [PubMed] [Google Scholar]
  25. Lipsett M., Wyant-Stein K., Mendes S., Berger E., Berkman E.T., Terplan M., Cioffi C.C. Addressing stigma within the dissemination of research products to improve quality of care for pregnant and parenting people affected by substance use disorder. Front. Psychiatry. 2023;14 doi: 10.3389/fpsyt.2023.1199661. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Long E.C., Pugel J., Scott J.T., Charlot N., Giray C., Fernandes M.A., Crowley D.M. Rapid-cycle experimentation with state and federal policymakers for optimizing the reach of racial equity research. Am. J. Public Health. 2021;111:1768–1771. doi: 10.2105/AJPH.2021.306404. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Mackie T.I., Sheldrick R.C., Hyde J., Leslie L.K. Exploring the integration of systems and social sciences to study evidence use among child welfare policy-makers. Child Welf. 2015;94:33–58. [Google Scholar]
  28. Mainstreaming Addiction Treatment Act of 2021, n.d.
  29. McCambridge J., Mialon M., Hawkins B. Alcohol industry involvement in policymaking: a systematic review. Addiction. 2018;113:1571–1584. doi: 10.1111/add.14216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. McGarity T.O., Wagner W.E. Harvard University Press; 2010. Bending Science: How Special Interests Corrupt Public Health Research. [Google Scholar]
  31. McGinty E.E., Goldman H.H., Pescosolido B., Barry C.L. Portraying mental illness and drug addiction as treatable health conditions: effects of a randomized experiment on stigma and discrimination. Soc. Sci. Med. 2015;126:73–85. doi: 10.1016/j.socscimed.2014.12.010. [DOI] [PubMed] [Google Scholar]
  32. McGinty E.E., Goldman H.H., Pescosolido B.A., Barry C.L. Communicating about mental illness and violence: balancing stigma and increased support for services. J. Health Polit. Policy Law. 2018;43:185–228. doi: 10.1215/03616878-4303507. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Mojtabai R. Estimating the prevalence of substance use disorders in the US using the benchmark multiplier method. JAMA Psychiatry. 2022;79:1074–1080. doi: 10.1001/jamapsychiatry.2022.2756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. National Academies of Sciences E. National Academies Press; 2019. Medications for opioid use disorder save lives. [PubMed] [Google Scholar]
  35. Nelson K.L., Purtle J. Factors associated with state legislators’ support for opioid use disorder parity laws. Int. J. Drug Policy. 2020;82 doi: 10.1016/j.drugpo.2020.102792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Oliver K., Innvar S., Lorenc T., Woodman J., Thomas J. A systematic review of barriers to and facilitators of the use of evidence by policymakers. BMC Health Serv. Res. 2014;14:2. doi: 10.1186/1472-6963-14-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Oliver K., Hopkins A., Boaz A., Guillot-Wright S., Cairney P. What works to promote research-policy engagement? Evid. Policy. 2022;18:691–713. [Google Scholar]
  38. Plante A., Lovell C. National Council for Mental Wellbeing; 2024. Lived Experience Safeguard Scale. [Google Scholar]
  39. Purtle J., LÊ-SCHERBAN F., Wang X.I., Shattuck P.T., Proctor E.K., Brownson R.C. State legislators’ support for behavioral health parity laws: the influence of mutable and fixed factors at multiple levels. Milbank Q. 2019;97:1200–1232. doi: 10.1111/1468-0009.12431. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Quorum, 2018. Meet Quor. World’s First Integr. Public Aff. Platf. URL 〈https://www.quorum.us/〉
  41. Rossow I., McCambridge J. The handling of evidence in national and local policy making: a case study of alcohol industry actor strategies regarding data on on-premise trading hours and violence in Norway. BMC Public Health. 2019;19:1–12. doi: 10.1186/s12889-018-6348-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Salem S., Weskott T., Holler A. Does narrative perspective influence readers’ perspective-taking? An empirical study on free indirect discourse, psycho-narration and first-person narration. Glossa J. Gen. Linguist. 2017;2 [Google Scholar]
  43. Scott J., Lubienski C., DeBray E., Jabbar H. Using Research Evidence in Education. Springer; 2014. The intermediary function in evidence production, promotion, and utilization: the case of educational incentives; pp. 69–89. [Google Scholar]
  44. Scott J.T., Larson J.C., Buckingham S.L., Maton K.I., Crowley D.M. Bridging the research–policy divide: pathways to engagement and skill development. Am. J. Orthopsychiatry. 2019;89:434–441. doi: 10.1037/ort0000389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Scott T., Pugel J., Fernandes M., Cruz K., Long E.C., Giray C., Storace R., Crowley D.M. Cutting through the noise during crisis by enhancing the relevance of research to policymakers. Evid. Policy. 2023;19:178–195. doi: 10.1332/174426421X16535828173307. [DOI] [Google Scholar]
  46. Scott J.T., Collier K.M., Pugel J., O’Neill P., Long E.C., Fernandes M.A., Cruz K., Gay B., Giray C., Crowley D.M. SciComm optimizer for policy engagement: a randomized controlled trial of the SCOPE model on state legislators’ research use in public discourse. Implement. Sci. 2023;18:12. doi: 10.1186/s13012-023-01268-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sigfusdottir I.D., Kristjansson A.L., Gudmunsdottir M.L., Allegrante J.P. A collaborative community approach to adolescent substance misuse in Iceland. Int. Psychiatry. 2010;7:86–88. [PMC free article] [PubMed] [Google Scholar]
  48. Sigfúsdóttir I.D., Thorlindsson T., Kristjánsson Á.L., Roe K.M., Allegrante J.P. Substance use prevention for adolescents: the Icelandic model. Health Promot. Int. 2009;24:16–25. doi: 10.1093/heapro/dan038. [DOI] [PubMed] [Google Scholar]
  49. Tseng, V., 2012. The uses of research in policy and practice. Society for Research in Child Development Washington, DC.
  50. Van Lissa C.J., Caracciolo M., Van Duuren T., Van Leuveren B. Difficult empathy: the effect of narrative perspective on reader’s engagement with a first-person narrator. DIEGESIS Interdiscip. E-J. Narrat. Res. 2016:5. [Google Scholar]
  51. Woolf S.H., Chapman D.A., Buchanich J.M., Bobby K.J., Zimmerman E.B., Blackburn S.M. Changes in midlife death rates across racial and ethnic groups in the United States: systematic analysis of vital statistics. Bmj. 2018:362. doi: 10.1136/bmj.k3096. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material

mmc1.docx (33.1KB, docx)

Articles from Drug and Alcohol Dependence Reports are provided here courtesy of Elsevier

RESOURCES